Since no one else is responding to this, I'll chime in.
First, the assumption that you are making about pyro stain acting as a "variable density low contrast filter" may not be true. This very issue has been discussed here at length (do a search). I was originally of the same opinion as you, but Nicolas Lindan, who probably knows more than any of us about pyro stains, has about convinced me that the color of the pyro stain is roughly equivalent to a grade 2 filter and does not have the effect of changing contrast appreciably between low and high values. So, what you have is a "variable density grade 2 filter" which may make some difference when printing with very high and low contrast filters, but otherwise will only have a small effect, if any. Although pyro negs may print differently on VC and graded papers, the differences in paper formulations, etc. prevent one from firmly attributing contrast differences to the pyro stain. Constructing a test for this is problematic.
What the stain does is add density that helps mask grain. If you make a mask as you describe, the pyro stain density would simply add to the density already in the negative, giving you more contrast, not less. I think you would be better off with a regular contrast-reducing mask than bothering with the pyro stain mask. If you do want to try the pyro stain mask, make a first-generation contrast-reducing mask, develop it in pyro and bleach away the silver. This will at least give you contrast reduction instead of an increase.
There are other techniques, which you are likely aware of that may help as well. You mention flashing, but seem to feel that the compression of values in the highlights is not appropriate for the goal you have in mind (however, this is precisely what you seem to be after with your "variable density low contrast filter"). Nevertheless, I find that small amounts of flashing in conjunction with other techniques quite helpful at times. Then, there are reduction techniques for the negative itself. These are irreversible, but have worked for many (I tend not to use them).
Another technique to explore is the "New Sterry Method," in which paper is bleached with weak dilutions of ferricyanide/bromide before development. This changes the contrast in the shadows and leaves the highlights mostly untouched. It works well and may be the thing you are looking for. I believe Unblinking Eye has some articles on this. A Google search will turn up lots of info.
FYI, bleaching the image silver from pyro negatives is easy. Just use a rehalogenating bleach of ferricyanide and potassium bromide. I use 20g potassium ferricyanide and 8g potassium bromide to 1 liter of water (or equivalent. The bleach can be reused.). I do this in conjunction with redevelopment in a pyro developer to add more proportional density (contrast) to a weak negative. The silver redevelops and more pyro stain is added, which gives the extra contrast. This is likely not suitable, however, for smaller-format negs, since the redevelopment changes the granularity of the negative. With LF negs, the effect is less noticeable.
Finally, however, the question arises as to why you are not simply working to get better-developed negatives that print better at normal contrast and don't need the kind of extensive masking you are doing. If you find you need to do a lot of masking to just control contrast, then maybe you should be developing less... Just a thought.