In the last fourteen years, functional magnetic resonance imaging (fMRI) studies in the field of high-level vision have identified areas of the human brain selective for particular categories, most notably faces (fusiform face area, FFA; Kanwisher et al., 1997), bodies/limbs (extrastriate body area, EBA; Downing et al., 2001), places (parahippocampal place area, PPA; Epstein and Kanwisher, 1998), and words (visual word form area, VWFA; Cohen et al., 2000). In this talk, I will focus on the fine-scale functional organization of face- and limb-selective activations. Previous research indicates that the location of the FFA and EBA is variable, both anatomically, as well as relative to other high-level visual areas. We have conducted a series of studies examining this variability in spatial location of the FFA and EBA using traditional stimuli used to localize these areas, but have implemented higher resolution fMRI (1.5mm voxels) for more precise measurements compared to past studies (3mm voxels). Results indicate that the variability in spatial location in previous studies is due to the fact that there is not one FFA and one EBA, but rather a series of localized face- and limb-selective clusters in distinct anatomical locations with specific functional boundaries relative to other known high-level visual regions. A separate experiment manipulating the visual field position in which the face or limb is presented indicates that these anatomically distinct face- and limb-selective clusters are functionally dissociable based on their position sensitivity. These results indicate that defining brain areas based on a single measurement of category selectivity is insufficient and I will conclude with a framework to consider before determining which functional activations should be considered visual areas.