The hydraulic properties of a fractured rock mass are largely controlled by connected fracture networks. A thorough understanding of the physical flow processes in fracture networks is essential for assessing the transport capacity of the rock mass. However, the fracture surface roughness morphology, fracture distribution characteristics, and fluid flow regimes strongly influence the flow capacity of a fracture network. To this end, the rough topographic characteristics of fracture surfaces were quantified using fractal theory, and then the effective permeability model and nonlinear seepage effect assessment model of the rough fracture network for different flow regimes were developed based on the possible occurrence of laminar and turbulent flows in a single fracture. Finally, the influences of the geometric parameters of the fracture network on the effective permeability and nonlinear flow characteristics were analyzed. The results show that the prediction results of the proposed models are in good agreement with the field test data and can effectively reveal the seepage influence mechanisms under different flow regimes. Additionally, the results show that the effective permeability is closely related to the fractal dimension, relative roughness, aperture scale, distribution characteristics, and hydraulic gradient of the fractures. The nonlinear behavior of fluid flow significantly reduces the effective permeability of the rock mass. The proposed models can provide a reference for evaluating the transport capacity of rock masses under different fracture distributions and flow regimes.