The pygmy angelfishes (genus Centropyge) provide numerous examples of discordance between color morphology, taxonomy and evolutionary genetic lineages. This discordance is especially evident in the Centropyge flavissima complex, which includes three primary color morphs, three previously recognized species (C. flavissima, C. eibli and C. vrolikii) and three distinct mitochondrial (mtDNA) lineages that do not align with species designations. Our previous research showed that the putative C. flavissima arose independently in the Indian and Pacific Oceans, and the three mtDNA lineages align with geography rather than species assignments. Here we add 157 specimens to the previous dataset of 291 specimens, spread across a greater geographic range, to pinpoint the distribution of mtDNA lineages and color morphs. We found that the mtDNA lineages show remarkably strong geographic boundaries corresponding to the Indian Ocean, Central-West Pacific and Central-South Pacific. We also test the validity of the ‘Black Tiger Centropyge’ in the C. flavissima species complex, a taxonomic oddity that is restricted to shoals and atolls off the coast of northwestern Australia, and the newly named C. cocosensis (Shen et al. sp. nov. 2016) assigned to the C. flavissima lineage in the Indian Ocean. We conclude that the Black Tiger Centropyge is not a valid species but an intermediate between sympatric color morphs that correspond to the putative species C. eibli and C. vrolikii. Our greater sampling effort also do not support the genetic distinctiveness of C. cocosensis given shared mtDNA haplotypes with the sympatric C. eibli and C. vrolikii, but instead we find conflicting lines of evidence concerning the taxonomy of this group. We urge caution and taxonomic restraint until the true nature of this species complex can be revealed.