Hm. Let me try to restate that to make sure I follow you.

Consider three categories of environments: (Er) real environments, (Esa) simulated environments that closely resemble Er, aka "ancestral simulations", and (Esw) simulated environments that dont't closely resemble Er, aka "weird simulations."

The question is, is my current environment E in Er or not?

Bostrom's argument as I understand it is that if post-human civilizations exist and create many Esa-type environments, then for most E, (E in Esa) and not (E in Er). Therefore, given that premise I should assume (E in Esa).

Your counterargument as I understand it is that if (E in Esw) then I can draw no sensible conclusions about Er or Esa, because the logic I use might not apply to those domains, so given that premise I should assume nothing.

Have I understood you?