Determining appropriate intervention methods for rehabilitation and restoration of historic masonry bridges requires understanding the true behavior of the structural system. The presented study addresses this need by focusing on nonlinear structural behavior of stone masonry arch bridges and investigates the effect of different constitutive material laws, modeling approaches, and multiple spans on simulating the true behavior of the structural system. An extensive experimental program was conducted on a ¼-scaled model of a selected span of the multi-span arch bridge consisting of the arch ring, spandrel walls, and the fill. The model was tested with and without the presence of an in-plane steel bracing system supporting the spandrel walls to study the effect of adjacent spans on the selected span. Line loading applied at the quarter-span was increased with constant increments until. Destructive experimental investigation was followed by numerical studies of the tested specimen by means of both macro and simplified micro modeling approaches with DIANA FEA software utilizing different constitutive material models including Multi-Directional Fixed Crack Model with/without Drucker-Prager plasticity material law and Total Strain Based Fixed Crack Model. Comparison of the experimentally measured values with those of the numerical model validates that both modeling approaches can accurately predict the collapse load and collapse mechanism. Although simplified micro model offers advantages such as capturing the observed local damage states and cracking at masonry joints between the units; the added requirements of time, computational effort, and detailed material knowledge to model such structures accurately, makes macro modeling technique more preferable for masonry bridges having a complex geometry.