Tissues such as the skin and mucosae are frequently exposed to microbial pathogens. Infectious agents must be quickly and efficiently controlled by our immune system, but the low frequency of naive T cells specific for any one pathogen means dependence on primary responses initiated in draining lymph nodes, often allowing time for serious infection to develop. These responses imprint effectors with the capacity to home to infected tissues; this process, combined with inflammatory signals, ensures the effective targeting of primary immunity. Upon vaccination or previous pathogen exposure, increased pathogen-specific T cell numbers together with altered migratory patterns of memory T cells can greatly improve immune efficacy, ensuring infections are prevented or at least remain subclinical. Until recently, memory T cell populations were considered to comprise central memory T cells (TCM), which are restricted to the secondary lymphoid tissues and blood, and effector memory T cells (TEM), which broadly migrate between peripheral tissues, the blood, and the spleen. Here we review evidence for these two memory populations, highlight a relatively new player, the tissue-resident memory T cell (TRM), and emphasize the potential differences between the migratory patterns of CD4(+) and CD8(+) T cells. This new understanding raises important considerations for vaccine design and for the measurement of immune parameters critical to the control of infectious disease, autoimmunity, and cancer.