Some examples seem excessively nitpicky. For example, "class" can be a generic word meaning "group" in addition to being a scientific term, so calling a phylum a "class" in a work intended for younger readers isn't inherently inaccurate.

I think there may be a justification for these:

• The ikran in Avatar apparently have the scientific name Pterodactylus giganteus. No. This is wrong. Being members of the genus Pterodactylus would mean that they're small pterosaurs from Earth's Late Jurassic period. For those not in the know, it's actually a four-winged dragon-like beast from a moon in the Alpha Centauri system. So yeah...
  • The Na'vi themselves are classified as Homo pandora, which would mean that they are closely related to humans. As we know they evolved from some sort of alien lemur creature, and not hominids transported from Earth, this is wrong.

From what I know, codes for Zoological and Botanical nomenclature allow homonym genuses between themselves. Maybe if alien species were discovered it would be valid to have homonym taxa with terrestrial species? Of course, this would be up to the discretion to taxonomists, but if it can be used for Earth species, the possibility of it repeating for alien species would render these examples invalid, in my opinon. What do you think?