Fair point BOAC.
Declaring an interest, I'm currently seeking funding (and perhaps a few suitably clever collaborators) to put together a research programme into use of AoA in primary flight guidance. A big job, likely to run on for several years, but should be worth doing if we can make it happen.
You'll always need IAS: as you say, for ATC separation (on jets, little interest for smaller aeroplanes), Vr, Vne, Vf. But, it should be able to schedule climb and approach more accurately using IAoA, as well as stall avoidance. Cruise may be able to be accurately scheduled with AoA also, but that's relatively unproven - the maths works, but there's no substitute for actually doing it.
G