Enhanced Speech - does it also match loudness and do other existing processing?

Trying out the new AI enhanced speech in Premiere Pro.

I'm a bit confused - after I press the button for Enhanced speech, and adjust the mix slider from the default 7 to my own choice (btw, why can't it simply be a 0% to 100% scale, instead of an arbitrary numeric scale),

Do I still fine-tune it with the additional sliders that have always existed for repair, clarity, denoise, etc?

Does enhance speech also do a loudness match, or should I still do that, if desired?

Does adjusting any of the existing "essential audio" sliders disable everything that enhance speech just did, or does it fine-tune it further?

When I continue to edit the clip (tightening cuts, etc.) Premiere will suddenly start re-procesing the enhanced speech again.  Is there a way to "bake in" the audio processing so it doesn't keep regenerating, but preserve the ability to fine tune it further?

I'm thinking of the classic workflow with Premiere Pro bouncing audio out to Audition and bringing it back with "render and replace" but it keeps shadow or side-car files so you can round-trip again to Audition later if desired.  I wish enhance speech would work with the same kind of nondestructive, but optimized workflow (I hope this kinda makes sense?)