Experimental autoimmune uveoretinitis (EAU) in animals can be induced by immunization with retinal antigens or their fragments and represents human uveitis of putative autoimmune origin. The pathogenesis of EAU, and likely also of human uveitis, involves cell-mediated destruction of retinal tissues that is dependent on retinal antigen-specific T cells. Because in most cases a Th1-type response has been implicated in pathogenesis, the prevailing consensus has been that immunoregulatory manipulations designed to enhance the Th2 response at the expense of the Th1 response will be beneficial in clinical treatment of uveitis. This assumption may not always be correct. The present review will summarize the evidence that, despite a central role for Th1 response in uveitis, an unopposed Th2-like response can be equally or more destructive to the retinal tissues. Furthermore, the Th1 response itself triggers regulatory circuits that feed back and dampen further recruitment of antigen-specific T cells into the Th1 effector pool. Thus, although the Th1 effector response can and does result in retinal pathology, immunoregulatory strategies must take into account that immune deviation therapies designed to replace the Th1 with a Th2 response might result in exchanging one type of pathology for another rather than in achieving the desired therapeutic effect.