Palm stump holes

[miamicuse]

This is kind of a continuation of a previous thread:

In that thread I removed cluster of Dypsis lutescents with a large root mound.

Since I had over twenty stumps around my property, some are palms and some are other trees, I hired a tree service to  grind all the stumps.

So here is before the Dypsis lutescents removal.

 

This is after the plant and roots removed to the surface.

This is after the stumps grinded down about 4 to 6 inches.

I have a few questions about the rest of the roots buried under.  There is still a massive root system below, and it's woody.  I hit it with a shovel and it's solid.  This root system will eventually rot out in a few years and turn mushy right?  If I want to lay down some brick pavers, is that going to work?  It will begin to sink as the root rots out and form kind of a crater right?

The next question is about the Roystonea regia.  It was about 50' tall and last year it was removed due to Ganoderma butt rot.  See the Livistona chinensis to it's left.

After the removal.

After the stumo grinding.

After the stump grinding was done I removed all the root shavings.  What I didn't realize was that when grinding the royal's stump, part of the Livistona chinensis roots got chewed up with it because they were so close together.

Here is a picture showing the damages done to the Livistona chinensis roots.  How concerned should I be?

Finally, there were some coconut palm old stumps that were already half rotted and were taken out.  This one hole is about 20" in diameter and over 10" deep in the middle.  I have a Beaucarnea recurvata (ponytail palm) that is about 8' tall 20" base diameter that would be a perfect spot for it.  It's not a palm so it will not be hurt by possibly having the ganoderma fungus there.  However, my question is if the root mass will continue to rot and decay, will this hole be stable enough for the Beaucarnea recurvata's new root system?  Or best to avoid these holes and dig new adjacent ones even though these new spots are not as optimum?