EDT, R T30, R T20

Recently acquired Fuzzmeasure for speaker in room optimization. It has become a very powerful tool for me. Speakers are positioned by using room optimizers. Room treated with GIK traps. Also currently experimenting with bagend bass traps. Frequency response is reasonably well optimized.

What I find is that 1-2 inches of speaker movements produce significant change in tonal balance not quite reflected in the frequency response. The tonal shift is explained in the change in reverb time in 1/3 octave: EDT, RT20, RT30. I read up on this and realize they are slope of different points of decay curve. They are related but do share a linear relationship.

Which of the above parameters is the most important in a 14x22x10ft room?

Is there such a thing as too dead a room or too short a time for EDT, RT30, RT20? Where is the sweet spot?

Interest is to produce orchestra hall like ambience. I do not focus much on Jazz or chamber.
My understanding is that they are different approximation of RT60.

EDT is the extrapolated RT60 time using decay time -5 to - 15db.

T20 is the extrapolated RT60 time using decay time -5 to - 25db.

T30 is the extrapolated RT60 time using decay time -5 to - 35db.

Is my understanding correct?

In my system the EDT and T20 are more representative of shift in tonal balance. Are these values the one I should be working on? The T30 of some frequencies are zero. How could that be?