A major roadblock to achieving a cure for human immunodeficiency virus type one (HIV-1) is the persistence of latent viral infections in the cells and tissue compartments of an infected human host. Latent HIV-1 proviral DNA persists in resting memory CD4+ T cells and mononuclear phagocytes (MPs; macrophages, microglia, and dendritic cells). Tissue viral reservoirs of both cell types reside in the gut, lymph nodes, bone marrow, spleen, liver, kidney, skin, adipose tissue, reproductive organs, and brain. However, despite the identification of virus-susceptible cells, several limitations persist in identifying broad latent reservoirs in infected persons. The major limitations include their relatively low abundance, the precise identification of latently infected cells, and the lack of biomarkers for identifying latent cells. While primary MP and CD4+ T cells and transformed cell lines are used to interrogate mechanisms of HIV-1 persistence, they often fail to accurately reflect the host cells and tissue environments that carry latent infections. Given the host specificity of HIV-1, there are few animal models that replicate the natural course of viral infection with any precision. These needs underlie the importance of humanized mouse models as both valuable and cost-effective tools for studying viral latency and subsequently identifying means of eliminating it. In this review, we discuss the advantages and limitations of humanized mice for studies of viral persistence and latency with an eye toward using these models to test antiretroviral and excision therapeutics. The goals of this research are to use the models to address how and under which circumstances HIV-1 latency can be detected and eliminated. Targeting latent reservoirs for an ultimate HIV-1 cure is the task at hand.