Interesting observation, and I agree with most of your findings. A few comments.
1. If you would like to test the base material of the film independent of the emulsion you can just strip the emulsion off and then measure the base. To do so just place the film in a solution of Clorox and after a minute or so you will be able to easily rub off the emulsion, leaving nothing but the polyester base.
2. There is a big difference in the transmission of the base of some films, especially in the UV range. Just compare the base of TRI-X with TMAX-100 in UV reading, for example, if you really want to be blown away.
3. To repeat what I have said before, which your tests show, is that proportional stain is the stain that is formed around the silver grains. It is very desirable and adds considerably to the micro-contrast of all films that are based on silver.
4. General stain affects the entire gelatin layer, regardless of whether there is silver density or not. It is nothing but extra fog density and serves no useful function, and under certain conditions is harmful to maximum image quality.
5. From the above one should conclude that staining developers work about equally well with all films that contain silver. That some films appear to work better than others is due to fact that the gelatin base of some films is thicker than that of others, and this films consequently develop more general stain, for the simple reason that there is more gelatin to stain.
6. Irrespective of anything you or I may say on the matter there will be many who continue to believe that a high level of general stain is desirable. But that should surprisse no one. Myths can be powerful things.