I have experience using EXPANDED PERLITE as filler in composite buildups - works fine. It's about 80% silicon dioxide (like glass) - the rest is mostly alumina. Quite inert under ordinary circumstances. Particles are typically larger than 100 microns, and the irregular/granular surface enhances interparticle bonding. Don't mix perlite with materials that react with alumina, unless you want that reaction.
Microballoon material (3M, for example) as filler is much the same, though it's nearly 100% glass and each particle has a smooth surface. The particles are so small (usually under 100 microns) that surface area is very high and therefore bonding is excellent.
BUT: Expanded perlite is cheap (check Lowes or Home Depot for a large but light bag) and Microballoons are expensive.
The only application I'm familiar with is as a thickening agent added to glue to make a putty/filler. Model airplane applications. My dad used to mix cabosil with two part epoxy, for much the same purpose.
From the web... If you added a couple of tablespoons of Cabosil to say, an ounce of epoxy, it would greatly thicken it, to the "Aeropoxy-ish" consistency. To get epoxy to the same consistency using microballoons would probably take 4 to 6 times as much microballoons.
Since Cabosil is a thickening agent, the resultant "goop" is stronger (and heavier, and harder to sand) than the equivalent "goop" made with microballoons. So, I use Cabosil where I need strength, like at the seam on a molded wing, and use microballoons where I intend to sand, like a fillet.
