This will probably never happen in this timeline for some sort of technical or economic (or political) reasons, but I feel like the Space Island Group (from the early 2000s) is relevant to this, and their counterpart in this universe would be more motivated to promote their concept of "commercial wheel-shaped stations made out of Space Shuttle external tanks":
I think if there is hope to achieve the kind of sortie rates and launch price reductions this sort of grand vision requires, ironically tank disposal (on a routine basis anyway) would be eliminated, because the kind of "STS Mark 2 or 3" required would be to integrate a reusable tank with the SSME descended main engines, along with relatively low cost many times reusable LRBs, into a fully recoverable Launch Vehicle, which would carry, probably still sidesaddle, an Orbital Package to very low Earth orbit, say 100 km (in the full range of inclinations). The Orbital Package would surely have to boost to a higher orbit, but orbital maneuvers are relatively modest in delta-V unless we are talking about transfers to GSO or to deep space, Luna and beyond. I vaguely estimate it ought to be feasible to give TPS to a tank (perhaps a heavier tank made of steel), put some maneuvering flaps fore and aft (the big ones aft since the engines are there). The uncrewed Reused Launch Vehicle would have to boost the whole mass to the very low orbit, which decays rapidly, but not I think so rapidly the separated RLV can't make many orbits to phase to a reentry arriving at a launch site. Perhaps overoptimistically guessing the RLV masses about 100 tonnes dry, the payload to very LEO is about 40 tonnes, which seems respectable--some OPs are just deliveries of heavy cargo, one-way with disposed (or orbitally repurposed) shrouds and engine/propellant requirements cutting the payload delivery to say 25 tonnes, others are reusable crewed Orbital Shuttles, where we design a bare bones Space Truck with crew escape capsule being an alternative to an uncrewed version with maximal payload, and Lego in a minimal escape capsule for say 8 crew we can use as ongoing mission (for a short mission) habitation for say 3 flight crew, amounting to 15 tonnes leaving 10 tonnes orbital payload, with downmass of 10 tonnes. Or a larger crew uses some of the payload capacity for an Orbital Habitation extension with an airlock and docking tower, and only delivers a handful of tonnes or a modest module.
Both RLV and a basic space truck Orbital Shuttle reenter the way Starship is supposed to, controlled belly flop leading to terminal "skydiver" horizontal descent, unlike Starship they don't flip into nose-up for final retro-propulsion engine landing, but rely on sideways (down) thrusters for a horizontal landing.
I've been hung up on replying to several posts back regarding the choice of oxygen rather than hydrogen peroxide to burn ethanol in the Keplers.
My thinking on those lines leads me to suggest the landing thrusters for the proposed RLV would basically be the +Z (in horizontal aspect) reaction control thrusters but cranked up to far greater chamber pressures than required in vacuum orbital maneuvering. This requires hyper-rating just those (4) thrusters to really high pressures. The orbital maneuvering engines generally operate at fairly low pressure (OTL Shuttle OMS and RCS at about 8.6 atmospheres) and have a fairly high expansion ratio of 55; this works great in vacuum for modest thrusts, but to land 100 tonnes of RLV at terminal velocity as high as say 100 m/sec will require a lot of oomph and very high pressure to overcome the high expansion ratio at sea level. But in a thruster cluster of four (one fore or aft on the X axis, one outward on the Y axis, two on the Z axis up and down) only one each needs to endure the superpressures reusably; if we have four such thrusters in terminal descent we can finely throttle at high thrust, we should have a well controlled plop-down. So we'd need to super-rate the pressurizing tank for an ethanol-peroxide system, or a much smaller pressurizing tank for just the ethanol in an eth-lox system along with super-strength for the LOX tank, and of course reserve propellant mass for the final landing, including pressurizing gas. It is basically the same sort of thinking to SpaceX's SuperDraco for the Dragon capsule. Versus the full mass of the RLV I suspect the special requirements just for landing are a modest element in the mass budget. Horizontal landing strikes me as better for ground handling--if were possible to land the RLV precisely at the actual launching pad as planned for Starship it might be a different logic to be sure. But SSMEs are gross overkill for a tail landing I suspect.
OK with that sort of approach, we no longer are expending ETs; that's part of the major point of it. We get high launch sortie rates because of rapid and inexpensive "inspect, tweak, gas and go" reuse of LRBs and ET/engine integrated RLVs, and the higher the launch sortie rate the lower the per-launch costs of NASA launch site infrastructural support; we have trained specialized highly paid ground crew working continually instead of sporadically.
But to get there we know in this TL there shall be Shuttle C's that only recover the engines. It should remain possible to make custom non-return ETs, with custom sidesaddle integrated payloads, and for instance launch a replacement for Enterprise's station core custom made. These ET's are built prioritizing their future long term mission in orbit and somewhat compromising on the initial propellant tank mission.
Clearly to get the drawing you provided we need, let's see, what is that, 30 ETs? 24 in the double outer ring and 4 more in a radial cross, which appears to be offset from the 24 tank rim pair by a tank along the axis of the 4 radial tanks spinning like a jack with another one being the axis of the big 24 tank ring and inner torus , plus God knows how you get the inner lower G torus, plus who knows how many cargo packet missions for the trusses and solar panels etc.
Also, I don't see how the solar panels are going to work like that, we'd be swiveling the heck out of them with each revolution.
It looks like the radius of the circumference the tanks lie end to end to form the ring is about two tank lengths, plus some allowance for core radius and tank extensions. Standard ET length OTL is just shy of 47 meters, but I'd propose if possible adding a couple to widen the intertank space for easy passage. This means more dry mass and might raise VAB clearance issues, but let's hope not. The very bottom of each ring tank would thus be at 102.3, and each tank centerline at 98 meters radius, plus allowance for the core radius which I am going to declare to be 14 meters for reasons that might become apparent, thus 112 to the centerline and half tank diameter or 4.3 more to the very outermost points. Taking the centerline as basis for nominal 1 G, spin speed is a hair over 33 m/sec, thus we cover one radian every 3.38 seconds and a complete revolution every 21.23 sec--2.83 revolutions a minute! Now that doesn't seem all that severe, and recent space station studies indicate it is probably entirely bearable for humans aboard. Also of course your pictures say nothing about whether achieving a full G on the ring is desired or not. It could be a Lunar G, about 1/6 Terran, or Martian G, about 1/3. I suspect though that realistic space station designs for long term habitation will need to offer at least some full-G habitation and would be surprised if anything much less than half a G offer great long term benefits (bad news for Mars colonists if so; perhaps Martian 1/3 G is OK, if people don't expect to return to Earth anyway, or for moderately short stays like say half a year). I am assuming a full G throughout though, just FYI.
Now I couldn't resist starting a long discourse on how I'd build such a station, in sequence, deploy the solar panels and heat rejection radiators, what sorts of population it might support (all on an interim, generally not permanent, basis, for reasons involving radiation hazards long term, also I suspect children and pregnant women would be banned outright, until such time as a fancier station design cuts the cosmic ray hazard down to levels at least some Earth populations are born, live full lives and die of old age in).
Let me boil it down to this:
a) the sensible station orientation as I see it, and likely orbit to be chosen for reasonable access to orbit and also radiation safety, such as we can achieve, will be such that both the Sun and Earth (and Moon, and the rest of the planets) are in roughly the same plane as the rings are in. Thus such objects would be "down" half the time and behind the arc of the station from any window on the spinning rings all the time, or pretty near. We don't want windows that look at the Sun! (Maybe for specialized solar astronomy--otherwise it is just a hazard). Therefore no windows on the spun elements. Well, making the station out of ETs largely puts the kibosh on that anyway. I believe one sensible design is to mount solar panels on the "floor" of spun elements, with this orientation--it means each panel delivers only 1/6 the average power it could if aimed properly at the Sun, where a factor of two in there involves the whole station being on Earth's night side, and the rest relates to half of it being in shade even on Earth's day side and to the sunlit half spending a lot of time oblique to the Solar direction. The panels each hung endwise on separate moving arms shown in the picture make little sense to me. I figure the panels installed on the "floor" will shade both the ring units and any radial ones connecting them to the hub and the hub itself, and this might mean thermal radiation panels might be few because black body emission of the shaded modules might be more than enough to keep them cool enough, even with considerable power consumption in each tank unit.
Therefore if the structure is a space hotel, dedicated 100 percent to paying tourists from Earth, their motive to shell out to go up there is presumably not to lounge around in little cabins at full G. They are either going to want to take advantage of lower gravity, or take a look at open space--where a low Earth orbit puts them into sunlight half the time. If we have a dedicated viewing cupola for purposes of looking at Earth, such a thing would be spun on the station axis at a different rate than used for artificial gravity, and a cupola facing Earth would need sun-shading only briefly and intermittantly. One on the opposite side of such a separately spun unit would on the other hand be exposed to heavy sunlight half the time, only allowing a clear view in the 45-50 min the station is shaded on Earth's night side. Aside from the starscape, only the Moon would show a disk, and look about the same as from Earth at night anyway. Relatively few people would want this view--whereas I suppose most everyone would want to spend a lot of time looking at Earth, day and night. Perforce this is freefall time.
The other things to do all involve either effectively zero or lower gravity.
Therefore 24, or even just 12, ET tanks for tourists are just mainly for them to sleep in, maybe other forms of familiar-gravity rustication, private space to get dressed in, have intimate private conversations, and stow their luggage. I guessed at a rather Spartan volume ration of about 12 cubic meters per person in the private cabins (bearing in mind space tourists will be rich people not accustomed to privations). OTOH, if the major purpose of the spun station is to provide people who are working long term, on a time scale of half a year to several years, in space, the major reason for the full G rings would be to encourage them to spend as much time as feasible there, "heavy time" to compensate for free fall adaptation on the job. So personal (and communal) volume in full Earth G would be more at a premium for them.
Allowing for the idea that even cramming the tourists into compact sleeping quarters to encourage them to get out to other parts of the station is the philosophy for a space resort, one has to allow volume for access, and communal spaces in full G (dining lounges for those who find learning to eat and drink in free fall or even reduced G a challenge, for instance) suggest that perhaps a 2000 cubic meter tank might be expected to house 100 tourists, while the same volume might only serve 50 long term space residents.
On that theory, your 24 ring module picture suggests a facility for 2400 space tourists--or 1200 long term space workers.
At a guess, such a large facility seems likely to be divided between both categories, so we have say 800 paying space tourists, and 800 working crew--say 300 of these are catering to the groundhog tourists one way or another, 100 are visitors from other orbital facilities that can't provide Earth G and have been rotated over for "heavy time," while the other 400 either maintain the station itself or pursue scientific work in low to zero gravity.
Now even that latter number, reduced as it is, is 50 times the projected sustainable population of SSE. Those 400 alone thus merit 50 ET--even considering SSE as described hitherto is not using 3/4 the ET tank, still 400 space science crew merit say 6 ETs. With the layout I envision, with just one big ring and not two, and eight radial tanks, half that number meriting 3 would have about 1/4 of 21, or 5, they more or less have a piece of.
To be clear the layout I think is realistic involves--at least one tank with long axis on the central station axis, the near free fall spindle, which is counterspun to turn one face to the Sun, serve as a mast for sun-tracking solar panels, and for spaceship docking--we might add more endwise, and then perhaps surround these with tight clusters of more free fall tanks around this longer mast. Centered on the minimum single spindle, we have a single strong central truss ring surrounding the central spindle. This has four tanks mounted endwise, radially that is, and each of these has a second tank on the end of the upper one. This creates the room for each arm of two to have two circumferential tanks hanging from their tips, leaving room for four tanks, each of which we can regard as split 50/50 between the spokes completing the circle of 12. Thus two of the four radial pairs are various G level recreation towers and corridors to the core spindle and thus tourist free fall zones, including the Earth viewing lounge cupola. (For mere access to the zero G core and mast, working crew also travel up and down the tourist towers, making a businesslike transit efficiently; as accesses the other two radial towers are backup for emergencies only, being primarily work spaces for biological researchers).
I cannot figure what the inner torus in the picture is for or how it was supposed to be made, but the secondary cruciform need not spin at all. It could be a bunch of free fall labs and workshops, Whereas tourists could do with another such, or say a bundle of six tanks wrapped around another central access/core tank, providing free fall cubic. If there are 800 tourists and 12,000 cubic meters of free fall play space for them, each one has 15 cubic meters there...which does not sound like much, but supposing each tourist spends only a quarter of their time in free fall, and basically coming from all over Earth are on three shifts so usage is pretty evenly spread out around the 24 hour clock, at any given time each has more like 60. Also, such a facility as an Earth gazing cupola-lounge would have to be in addition to whatever we fit inside ETs.
We might need more intermediate gravity facilities for both tourists and scientists. If one purpose of this station is to be a jumping off point for Lunar expeditions we might want extra room at 1/6 G (hence radius under spin) to prepare crews headed there, or to ease crews returning from long Lunar visits before moving them into full Earth surface G facilities. If we are staging for Mars, we'd want more prep modules for outbound crews, and tourists might well want to spend time in reduced but non-zero G.
Therefore I question why this design has doubled up on the full G rings but seems a bit scanty on variable G or free fall volume.
Of course, the thing could be a staging base for a massive emigration push to Mars for instance, and all 24 of the ring tanks could be for acclimating prospective colonists to Mars G, giving a chance to scrub out any individuals who prove they cannot safely adapt before committing them to a ship headed that way.