-Well it was the Byzantine that ousted the Vandals.
Not exactly : Berbers really did a number on Vandals, threatening to swallow most of Roman Africa at their benefit, which frightened Africano-Romans who called Constantinople to help : Byzzies crushed Vandals but (due to poor geostrategical planning) were stuck in a "low-intensity warfare" with Berbers for decades, ending up with a smaller coastal territory than Vandals ever had.
-Well as far as I know the Asturian at times stressed their Visigothic heritage, despite their region being one of the most peripherical ones. It did never reach the point where it could have overtaken Iberian culture and language though.
Gothic, since the VIIth century is essentially the same than what Frankish meant for northern Gaul at the same time : a political identity over a Romance culture.
Could an alternate 3rd century to 5th century period lead to possible Germanic take over that is more organized?
If anything it would make them probably less organised : the ethnogenesis process that led to the creation of "border peoples" as Goths, Franks, Alamans,etc. was in effect, but these peoples barely began to exist as such. All the structurating proess came from (bilateral or forced) exchanges with Rome on an economical and political level, which is really blossoming (at lest for Barbarians) from the IIIrd century onwards. At this point, while "Gothic" or "Frankish" labels did existed, they were still based on various tribal entities (at least 4 or 5 distinct tribes).
Now, I do think it would make them even more easily swallowed up by Romania, culturally and institutionally, but I'm trying to answer the OP not on realistic or more or less realistic grounds, but following his general idea even when definitely implausible there.
Does the romanization required for a cohesive Germanic state to form prevent any kind of linguistic or even cultural assimilation? I'm speaking of regions like Northern France and Northern Italy, Iberia and Peninsular Italy are obviously off-limits.
Think of romanisation as a multi-faceted (not just cultural, but institutional and material), creolie-ising process. Britto-Romans were romanised,in spite of a limited latinisation : latinisation itself wasn't a decisive marker of romanisation as it was made in Eastern Romania on an hellenistic background. A continental superpower as Rome, compared to tribal and post-tribal chiefdoms IS bound to influe on them importantly, even with a limited acculturation.
Note that northern Gaul and Italy were very importantly romanised on most of these regards (if anything, Rhineland might have been more latinized than, say, Brittany, because of the importance of the population and the army), and it really required an absence of post-imperial features (generally outside Gaul itself) and the srength of Germanic-speaking centers (always outside Gaul) in late early Middle-Ages to undergo a linguistical germanisation.
Similarily, the germanisation of small parts of Northern Italy is something you can't really trace bacl to early medieval period (not because you didn't have isolates, but the linguistical border can't really be traced, we're talking of probably much more mixed geolinguistical situation) and stabilisation of linguistical geography really happens in the Xth to XIth centuries.