Basin-range coupling is a key process for energy and material exchange within the continental lithosphere, fundamental to understanding plate tectonics, resource distribution, and geological hazards. Its related physical modeling research has undergone a notable evolution from geometric similarity to dynamic similarity, and from single tectonic simulation to multi-process coupling. By reviewing the development of this field, it is pointed out that the differences in stage division are essentially driven by theoretical breakthroughs and technological innovations. Regarding experimental materials, modeling materials have evolved from basic substances such as quartz sand and silicone to composite functional materials capable of accurately matching the rheological behavior of geological prototypes. This paper further compares the density, friction angle/viscosity, and key dimensionless parameters of brittle, ductile, and plastic materials, and discusses the differences in material behavior under hypergravity conditions as well as the requirements for dynamic similarity. Based on the type of driving force, experimental models can be classified into three categories: compression, extension, and strike-slip. These models respectively elucidate the dynamic mechanisms of foreland basins, rift basins, and strike-slip transfer zones, consistently highlighting the controlling roles of pre-existing structures and rheological layering. By comparing three representative basin-mountain systems under the same dimensions, this paper reveals the complementarity between parametric experiments and case studies, along with the inherent limitations of physical modeling. Currently, the integration of cutting-edge technologies such as hypergravity centrifugation, CT scanning, digital volume correlation, and numerical modeling is driving a profound transformation of physical modeling towards real-time, three-dimensional, and quantitative analysis. Addressing the limitations of scale effects, time compression, laboratory apparatus influences, and lack of thermal effects, this paper proposes four technological development directions: high-resolution observation and fine modeling, intelligent continuous monitoring and parameter inversion, full-process multi-field coupled simulation, and multi-source data assimilation. Accordingly, a closed-loop research paradigm is constructed: “3D printing of initial models—hypergravity centrifugation for stress environment—digital volume correlation for four-dimensional strain field capture—artificial intelligence for automatic extraction of structural elements—numerical modeling for parameter inversion and evolutionary trend prediction”. In the future, this field will pay more attention to full-process simulation of “tectonics-geomorphology-sedimentation” and deep integration of artificial intelligence recognition technologies, thereby providing critical experimental support and theoretical frameworks for a deeper understanding of the operating mechanisms of Earth’s dynamic systems.