> is that Kant does not specifically depend on Newtonian assumptions but that human perceptions have to be mediated by time and space
I understand the point about mediation, and I tried to speak to it directly in the portion you quoted. You are exactly right about the way Kant uses that in his argument. It's flexible enough that it can be understood as resilient, I think.
I tried show how science can nevertheless speak to it in several ways. One is the convenient equivalence between Kant's necessary conditions and nature as we understood it at the time. That equivalence makes it easier to be comfortable positing time and space as we understand them as necessary conditions that are prior to nature. It would be harder (but not impossible!) to entertain if there was daylight between what we believed our faculties to be and what we believed the natural world to be, if what was "necessary" was something different than what we thought we experienced (such as, say, a singular loaf of unified spacetime). An implication of Kant's view, as I understood him, was that we should be ready to believe it's different anyway, but nevertheless, it's easier to swallow when it perfectly aligns with how we thing the world really is.
Second, for a different way of stating a similar point, if time and space are non-newtonian in some important way, we may have to understand that that is what we experience in the first instance, and we would have to wrestle with the fact that our intuitions may attest to a variety of possible underlying necessary conditions. (Wittgenstein has been mentioned in this thread, so for a paraphrase of something he supposedly said, in reply to a student noting that it looks like the sun goes around the earth: "What would it look like if the earth went around the sun?" So it may be with the nature of our faculties.)
Third, as we gain deeper understanding of brain functions of ourselves and other creatures, we may have inroads to how certain of our conscious experiences depend on conceptual abstractions or models. And if so, it transforms it into an open empirical question rather than one where we are simply stuck presuming necessary conditions. Kant might insist that that is all already through the lense of our faculties such as they are, but we, with more information, would be increasingly comfortable just not entertaining them as we get increasingly robust empirical understandings of conscious experience.
None of these is a "defeater" argument by itself and there's a lot that can be talked about here of course, but I think given 21st century science, if Kant were trying to present his vision for the first time we might find it more challenging to reconcile with what we know of the natural world.