The extensive mineralization of Cenozoic deep-sea ferromanganese deposits is a unique phenomenon in geological history. Ferromanganese nodules/crusts have attracted considerable attention owing to their enrichment in critical metals. The colloidal genetic model of hydrogenetic Fe-Mn mineralization has been widely accepted and applied since it was first proposed in the mid-1990s. With the rapid development of nanogeoscience over the past two decades, it has become clear that nanoparticles, as the smallest part of colloids, can significantly affect the Fe-Mn mineralization process owing to their unique properties. It has not only been discovered that iron and manganese oxides generally coexist as nanoparticles in various supergene geological environments, but it has also been verified that the primary Fe-Mn minerals, such as vernadite and ferrihydrite, in hydrogenetic ferromanganese nodules/crusts are nanoparticles. Iron oxide nanoparticles can catalyze the surface oxidation of Mn(II), thereby potentially explaining why hydrogenetic Fe-Mn minerals are usually symbiotic even at the nanoscale. In addition, Mn(III) minerals, which have generally been neglected in previous research, have been abundantly observed in ferromanganese crusts. The Mn(III) fraction is highest near the surface of the crust and gradually decreases with increasing depth, whereas the Mn(IV) fraction shows the opposite trend. The surface energy variation among manganese oxide nanoparticles with different valence states induces the initial precipitation of Mn(III) minerals during Mn(II) oxidation, which may eventually be converted to Mn(IV) minerals over geological timescales. Further progress in comprehending the Fe-Mn cycle in seawater and Fe-Mn mineralization on the seafloor through the advancement of nanogeoscience and high-resolution in situ experimental techniques is expected in the near future.