Hi again Al,
I've only used the hot water method to heat the material once, since it tends to be a bit of messing about in my opinion.

One approach when using this method, is to have some form of hook to hand and clump the 'fish eggs' together with it, then pull the clump out of the water and squeeze it to force out the water. A simple piece of bent wire would suffice. I don't imagine you could get all of the water out using this approach and thus, I moved to using hot air.
It would seem from your own observations using water at 200°F, the material takes quite a long time to cool and the thicker the 'part' dimensions, the longer the cooling period. 200°F / 93°C appears to be above the temperature I have been using, since the consistency for me seems more firm like modelling clay, such as Playdoh, rather than wet clay.
Using hot air, such as my butane torch with heat-shrink tip, or a SMD rework station, or even a heat gun, as soon as the granules/beads/fish eggs become clear, I knead them together and start to form whatever shape I need. If I am a little slow and the material starts become opaque, or the 'part' is a little intricate, I can use a little more hot air to soften it again. You could, if using the hot water method, dip the 'part' in again to soften and rework it though.
I guess it basically comes down to what you have to hand. It seems like using water at the temperature you used gives you a higher working time and a more malleable consistency, coupled with absolutely no risk of reaching the temperature at which the polymer starts to degrade, but with the added effort of having to squeeze the water out first. The flip side, when using the hot air, runs the risk of either blowing the granules/beads/fish eggs around in the dish due to too much air flow, or potentially reaching the temperature at which the polymer degrades. I suppose that's why I keep a good distance and remove the heat as soon as the material goes transparent.
If you use a kettle though, you also have the opportunity to make a cup of tea or coffee too.
Your good and bad points are pretty much spot on, but if you take GOOD #4 into consideration, BAD #2 becomes less of an issue.
1. How easy does it get contaminated with another substance.
2. How well does it hold up to various products like gasoline, alcohol, etc.
3. How well does it take paint, and what kind of paint.
4. How well it glues together, or what kind of glue to use.
5. Havent tried drilling it or tapping it, but i would think a very slow speed would be needed.
1. The flex shafts and adapters in the previous pics have been handled with dirty oily hands. You can see that the material is now not quite the same nice white is was from the tub. In addition, I know it's not good for either myself or the PC, but I tend to smoke quite a lot at the PC and I have a small DIP-8 to SOIC homebrew stripboard adapter for EEPROMs and such, which got 'tidied up' with some Polymorph. It sits infront of the PC and has yellowed somewhat over the years due to smoke being drawn over it by the PC fans. I haven't tried yet, but I believe that colouring agents can be used. There is a Youtube link here:
https://www.youtube.com/watch?v=Po8BaYD9Wks
2. I cannot comment as I haven't tested this.
3. Again, I haven't tested this.
4. The best thing about this is, you don't need any glue. Simply locally heat two 'parts' where you need to join them and press them together. The flex shafts were made this may. The shaft itself was
roughly formed and then rolled between two pieces of glass. Two rods were used as spacers, with the polymer inbetween. The 'knurled' handle was made by pressing the soft material into a former, which was actually the annular gear ring from an electric fuel pump. Once both parts were made, the centre of the handle was heated along with one end of the shaft. When both parts started to become transparent, they were mated and allowed to cool. The 1/4" square drive on the shaft was made in a similar fashion - the end was heated and pushed into the female square drive of a regular 1/4" socket, used as a former.
5. I too have not tried yet and agree that a very low speed would be necessary, but have just thought about creating threads... Why not mould the material around the threads of a bolt whilst making the 'part', then remove the bolt later and clean up where required?
I hope I've covered most, if not all, of your points and observations.
I do think this is one of those 'wonder' products, where you wonder how much more work/time/effort would be involved when using something else...
Regards.