We show the conclusions claimed in the manuscript arXiv:1202.5309v1 by Cuoco,
Komatsu and Siegal-Gaskins (CKS) are not generally valid. The results in CKS
are based on a number of simplifying assumptions regarding the source
population below the detection threshold and the threshold flux itself, and do
not apply to many physical models of the blazar population. Physical blazar
population models that match the measured source counts above the observational
threshold can account for 60% of the diffuse gamma-ray background intensity
between 1-10 GeV, while the assumptions in CKS limit the intensity to <30%. The
shortcomings of the model considered in CKS arise from an over-simplified
blazar source model. A number of the simplifying assumptions are unjustified,
including: first, the adoption of an assumed power-law source-count
distribution, dN/dS, to arbitrary low source fluxes, which is not exhibited in
physical models of the blazar population; and, second, the lack of blazar
spectral information in calculating the anisotropy of unresolved gamma-ray
blazar emission. We also show that the calculation of the unresolved blazars'
anisotropy is very sensitive to the spectral distribution of the unresolved
blazars through the adopted source resolution threshold value, and must be
taken into account in an accurate anisotropy calculation.