Abstract: To address the challenge of balancing computational efficiency and simulation accuracy in urban flood modeling, a new method coupling a virtual storage network with the Storm Water Management Model (SWMM), termed StoSWMM, is proposed. In this method, a virtual storage network is constructed within the SWMM framework using subcatchments as basic units. Bidirectional water exchange within the surface system and between the surface and drainage system is simulated through weir structures, while surface water depth is efficiently estimated based on the difference between virtual storage water levels and DEM elevations, enabling full-process flood simulation. The Jiangning Campus of Hohai University is selected as the study area, where StoSWMM is compared with a two-dimensional hydrodynamic model (2DHM) and a coupled hydrologic–hydrodynamic model based on nodal overflow (CHHM). Results indicate that, under the study area and designed rainfall scenarios, StoSWMM shows good agreement with 2DHM in terms of inundation extent, maximum water depth, and flood evolution. In areas with maximum water depths exceeding 0.3 m, spatial correlation coefficients range from 0.86 to 0.93, whereas those of CHHM range from 0.07 to 0.30, reflecting differences in surface inundation representation mechanisms and spatial conceptualization among the models. Under identical hardware conditions, StoSWMM achieves second-level computational performance, with runtimes approximately 1/40 of CHHM and 1/300 of 2DHM. Fully built upon the open-source SWMM platform, StoSWMM maintains physical consistency while achieving high computational efficiency, providing a lightweight modeling approach for rapid urban flood simulation and scenario analysis.