SLF here, please be gentle!
Too many pages, not enough time, but it seems to me that the MCAS is functioning solely on inputs about the attitude (AoA) of the airframe in one direction only. No attention appears to be paid to the position of the airframe in space - i.e. is altitude decreasing? AoA says nose is high but longitudinal gyro disagrees and says flight is level? Inertial Nav Systems have been around for a long time - back when I worked in the aviation industry. OK, that might be too expensive, but my yacht had a gyrocompass that cost about $100 as part of the autohelm. Why are these simple technologies not used for cross-checking of something mission critical like AoA? Before pushing the nose down, it would be helpful to know where it was pointing beforehand. And how could an automated system be allowed to drive the nose down while the altimeter shows altitude decreasing and acceleration increasing - the end result is inevitable!
It is not as easy, remember there are a bunch of certification rules that basically require the algorithms to be simple enough and predictable enough.
The altitude does not change anything to the problem, you could be stalling when recovering from a descent if you pull the stick too much, in that case the airplane is still in descent. So you have to rely on the AOA. You cannot rely on the gyro because the angle of attack and the attitude are not the same thing, for example if you get a sudden strong updraft your angle of attack increases but your gyro does not show that change.
It is a complex problem. There are no easy solutions.