Actually, while I'm at it (and I'm putting this seperately because in general it's a seperate point), a potted version of the history and mythology of Magnus Maximus/Macsen Magnus. Where to draw the line between the two is really quite difficult.
What is certain is that there was a man called Magnus Maximus, who came to Britain with the army, most likely transferred from Spain, and who had served for some time and become a reasonably high ranking official who had possibly served in a minor role during the Great Conspiracy. At some point he married, and had a son, Flavius Julius, and an unknown plural of daughters. Having gained the loyalty of the army in Wales, they declared him Emperor in 383, wheruppon he led the army to Gaul- marking the last known point of Roman troops in Wales- established a capital at Trier and at one point forced Valentian II from Rome. He would be killed in battle in 388AD, and Flavius Julius strangled when Trier was recaptured. It is known that Magnus' Wife was definately around in Trier, and appears to have been spared, while more concretely both his unnamed mother and the unknown number of daughters were spared. One daughter may have married Ennodius the Proconsul of Africa.
The link with Macsen Wledig is pretty secure- the dates and events map out well enough- and it is in this form that we see the Pillar of Eliseg site him as one of the early ancestors of the Kings of Powys through another daughter Sevira who was supposedly married to Vortigen. Traditional Welsh history/mythology also gives him two wives, Ceindrech ferch Reinden who is the mother of both Flavius and Eugenius (Owain Finddu) an unverifiable figure who was supposedly the first 'King' (probably governor) of what would become Morgannwg/Glamorgan but at the time is referred to as 'mid-south Wales'. He then marries a second time, this woman being St. Helena of the Host, or Elen Lwyddog, supposed daughter of Octavius/Eudaf Hen, Cheif/King of the Gewissae in Gwent and the wife who would meet with St. Martin of Tours while in Gaul and would bring Celtic monasticism to Wales. There's a whole lot of tangled mythology, history and hagiography in here, but the essential picture is that they have 5 children: Antoninus Donatus/Anwn Dynod, the King/Governor of South-west Wales (early Dyfed essentially), Constantine/Custennin Fawr the King/governor of North-west Wales (basically early Gwynedd), Publicus who would become a major monasticist under the name St. Peblig of Llanbeblig, and two daughters Gratianna who would marry the King of Dummonia (and features in at least one genealogy of King Arthur) and Severa, the wife of Vortigen. Pretty much every one of the later kings of Wales traces his descent through one of these lines (Peblig excluded, he supposedly remained celibate, while no-one traces descent from Flavius).
Where myth and reality merge is that we know the Romans essentially retreated from Wales with Magnus, and that the result was a mixture of traditional local chiefs asserting authority and Roman/Romano-Brtish governors becoming de facto chiefs in this area. We know that Magnus's wife took counsel with St. Martin of Tours, and so St. Helen who also supposedly did so at the same time seems a good fit for her. It seems reasonable to assume that Magnus would have left the governance of Wales in the hands of his younger sons, particularly if he'd married into the local tribal royalty. The fact that his daughters basically disappear from historical record lends fuel to all of this of course, as does his unnamed mother cropping up in the 380s. The link with Crispus/the baby of 322 being his father generally seems to be due to the fact that both disappear from the record soon after 326, and Magnus's parents are similarly missing from the historical record.
It's all a lot of conjencture and some pretty big bits basically have to be taken on faith, but it does at least hold up as a reasonable string of events and nothing comes across as completely impossible.
To link this back to the OP and the question, how to interpret this becomes a matter of if you would consider it reasonable that Macsen/Magnus is descended from Crispus, and that he had a pretty important role, possibly through children, possibly through appointments, in how post-Roman Wales would end up being divided. If so, then we can quite possibly write off the whole of the current set up entirely- Gwent may swallow Glamorgan, Powys could easily end up filling the void in Gwynedd and Dyfed, the Irish could come earlier and take more without a more clearly defined polity to take over in what is now Pembrokeshire and Caernarvonshire. If not, this is really just an annecdote and butterflies will probably mean lots of changes anyway.