There are many rodent models of autoimmune diabetes that have been used to study the pathogenesis of human type 1 diabetes (T1D), including the non-obese diabetic (NOD) mouse, the biobreeding (BB) rat, and the transgenic mouse models. However, mice and rats are not humans, and these rodent models do not completely recapitulate the autoimmune pathogenesis of the human disease. In addition, many of the reagents, tools, and therapeutics proposed for use in humans may be species specific and cannot be investigated in rodents. Researchers have used nonhuman primates to more closely mimic the human immune system and, to study species-specific therapeutics, but these studies are associated with additional ethical and economic constraints and, to date, no model of autoimmune diabetes in this species has been described. New animal models are needed that will permit the in vivo investigation of human immune systems and analyses of the pathogenesis of human T1D without putting individuals at risk. To fill this need, we are developing humanized mouse models for the in vivo study of T1D. These models are based on our newly generated stock of NOD-scid IL2rgamma(null) mice, which engraft at higher levels with human hematolymphoid cells and exhibit enhanced function of the engrafted human immune systems compared with previous humanized mouse models. Overall, development of these new generations of humanized mice should facilitate in vivo studies of the human immune system as well as permit the investigation of the pathogenesis and effector phases of human T1D.