This paper describes a new set of results on visual search of displays of 75 objects that differ in size, shape, and color, and presents a cognitive architecture model based on the active vision concept that accounts for the effects using object eccentricity and size effects, noisy saccades, and fixation memory provided by a persistent visual store. The data confirm older, less complete studies of this task. The model is a significant refinement of earlier visual search models and preliminary fits show that it promises to provide an integrated architectural account of these effects.