I live near Porland, OR, Matt; and yeah, I've got lots of clay in the yard. I will get to that, but right now I'm interested in pushing the low thermal inertia and low labor aspects, along with low cost. While these things aren't practical at all as is, I think I'm making a useful contribution to the communal knowledge base. After all, when I first started asking, "why not use Roxul," I couldn't find a definitive answer. Now we know.
By the way, FWIW, someone on a DIY forge or pizza oven thread claimed Roxul was usable up to about 600˚C=1100˚F. Sounds about right, but I don't have a link and don't have any handle on the writer's rep, except he sounded knowledgeable.
I think the backer board keeled over pretty early, going by the heat mark difference between the two sides. I'm not sure it was the main problem, though. I will know more when I get back home and have the time to dissect BOOH II more thoroughly. I need to cut through some pieces to make sure that the marks aren't just at the edge gaps. I'm sure I could get it to run better by putting a shroud on the bottom and venting to a chimney, of course.
For BOOH III, I will be investing in some Kaowool or equivalent. At a minimum, I expect to be able to make a system that is less leaky, isn't so vulnerable to structural collapse, and doesn't burn out at less than 2000˚F.