Though the development of human-to-mouse xenotransplant models is in its infancy, astonishing progress has been made in a short period of time. Two experimental applications have been developed: short-term transfer of human lymphocytes to generate models for autoimmunity and infectious diseases, and long-term engraftment of tissues with self-renewal potential. Human PBL-SCID mice have been used by multiple laboratories to study normal and autoimmune antibody responses, and have been shown to be readily infectable with HIV-1. SCID mice grafted with fetal tissue have been developed for studies of HIV-1 infection and its therapy as well as for studies of human hematopoietic cell differentiation. Human tumors appear to grow better in SCID mice than in nude mice, and hu-PBL-SCID mice can develop EBV-related B cell lymphoproliferative disease that resembles the immunoblastic lymphomas appearing in immunosuppressed transplant recipients. There is some evidence of mouse NK cells responding to the human xenograft, and of human T and B cells responding to mouse xenoantigens in these models, but these responses are not generally strong enough to have a major impact on human immune function. The use of these surrogate human models is expected to have a major impact on the understanding and treatment of human disease.