The problem is that "simultaneity shift" is really not possible. You can't go instantly from zero to 0.8c. Of course it works at any acceleration and final speed, but then it is less obvious that time dilation isn't caused by acceleration. I think ynot needs a leg up first before he tackles this, but I could be wrong.
That stuff about how infinite acceleration isn't possible, is irrelevant, but you know that already. You may worry that others might not understand this, but I think it's pointless to worry about that, because anyone who can't get past a detail like that, has a lot to learn about physics in general before it makes sense to even
begin to try to understand relativity. Newton's laws would be a better starting point.
I understand why you'd like to eliminate acceleration from these relativistic problems: You have noticed that some people incorrectly conclude that acceleration must be the cause of some "weirdness" in e.g. the twin "paradox" problem, and you have found a quick and easy way to prove them wrong. It's a good thing that you're able to prove them wrong without actually solving the problem, because now they don't have to understand the full solution to see that they're wrong. However, what you've said
only explains why acceleration isn't a relevant part of these problems, and why A and B will be the same age.
That A and B will be the same age is implied by 1, and ynot already understands that. He understands both 1 and 2. His problem is that he thinks that 2 contradicts 1, and the only reason for that is that he doesn't understand 3.
3 is the reason why 2 doesn't contradict 1, so there's no way that anyone can understand this problem without understanding 3. (And that's true regardless of whether they consider ynot's formulation of the problem or yours).
1. The scenario is perfectly symmetrical, so any correct statement abut A and B will still be correct if A is replaced with B and vice versa. (This of course includes 2 and 3 below).
2. In both of A's frames, B is aging slower than A. (This is time dilation).
3. When A turns around, he goes from a frame where B is younger to a frame where B is older. (This is not time dilation, it's relativity of simultaneity).