It still disturbs me that you have two hundred threads... but, since you have specifically excluded consideration of this aspect of the system in your OP, I suppose that it cannot be reconsidered.   But if it could...

The Windows memory-manager is known to be quite “lazy” and sometimes this characteristic produces memory-consumption that is quite larger than you intuitively suppose it should be.   I also suspect that it over-states the “amount used.”

My instincts still tell me – nay, they fairly scream at me – that you are going to have to find a way to reduce the total number of threads that are active at one time.   It has consistently been my experience over many years that a limited (and fairly small) number of threads should be put to work against an arbitrary (and perhaps quite large) number of things-to-do.   In the Windows environments, this seems to be particularly important.