Abstract: In recent years, extreme rainfall events such as the "21·7" Henan and "23·7" Beijing–Tianjin–Hebei storms have frequently impacted North China and the Huanghuai region. These events exposed significant gaps in our understanding of their physical mechanisms. This article summarizes recent researches on spatial-temporal features, synoptic patterns, mesoscale convective systems (MCSs), and microphysical structures of extreme rainfall. By combining multi-source observations and model simulations, the study highlights the multiscale processes—from large-scale circulation to microphysics—that jointly shape rainfall intensity. A multiscale pattern recognition method is proposed to improve the identification of extreme events. Two distinct MCS types—warm-type and deep-convective-type—are characterized, especially in terms of raindrop spectra and hydrometeor composition. A multiscale conceptual model tailored to North China is presented, offering insights to improve forecasting and model representation of extreme rainfall processes.