Animal models of human type 1 diabetes will be of a great importance for the evaluation of new combination therapies with curative potential. However, reliable predictive power for successful translation to patients with type 1 diabetes is crucial. This will be particularly important in the future when evaluating success of new combination therapies that show great promise for preservation and restoration of beta cell mass and thereby reverse the type 1 diabetic hyperglycaemia. But not all spontaneous animal models are equally well suited for this purpose. The advantages and disadvantages of the three spontaneous rat models (BioBreeding diabetes-prone [BB] rat, Komeda [KDP] rat, and LEW.1AR1-iddm [IDDM] rat) as well as the NOD mouse, compared with the characteristics of human type 1 diabetes, are considered in this review.