The question is not whether they can "technically" do it - of course they can, as they've already demonstrated. The question is whether they can make it work from a market perspective.
Last time they transitioned from a niche architecture to the architecture that was already mainstream - and happened to be faster and more power-efficient as well. That was basically a no-brainer, it could not fail from a market perspective (from a technical perspective however, it could), because it essentialy tore down previously-existing compatibility barriers between whatever software the mainstream of computer users used and what Mac users used. This time, they would transition AWAY from the mainstream laptop and desktop computing architecture to a niche architecture in that area, which, while still maybe being more power-efficient and maybe even faster (we'll just have to give them the benefit of the doubt there until we hold actual hardware in our hands), erects new compatibility barriers between the mainstream architecture and all people using software based on it and Apple's computer offerings.
This can only work from a market perspective if they either are big enough to drive such a momentum of customer and developer attention to this "new" platform that they essentially manage to make it successful all by themselves, or if they actually happen to be the harbinger of a future movement to ARM in desktop computing scenarios which will at some point in and out of itself be much larger than Apple's share of desktop computing space.
If either one of these happens to turn out to be true, they will be able to make that transition work. If none of these turn out to be true, they will fail (measured by falling Mac market- and mind-share), regardless of whether they execute the transition well from a technical standpoint, which I agree they will be very likely to do.