We introduce texture spectral criteria, which allow us to predict whether simplified spectrally factorized random field-based texture models, a set of two-dimensional models, can faithfully replicate texture spectra compared to their fully spectrally correlated 3D counterparts. These probabilistic models incorporate essential two- or three-dimensional building factors for modeling the seven-dimensional Bidirectional Texture Function (BTF), the most advanced representation in real-world material visual properties modeling. While these models seamlessly approximate original measured massive data and extend them to arbitrary sizes or simulate unmeasured textures, evaluating typically involves time-consuming synthesis and psycho-physical evaluation. The proposed criteria provide an alternative approach, enabling us to bypass the spectral quality evaluation step.