In recent decades, scientists have developed a wide array of hydrological, hydrodynamic, and other models to understand the dynamics of the Mekong River Basin. Indeed, the area has been described as 'flooded' with models. Drawing on STS and the philosophy of modeling - which has described models as mediating instruments - the first half of this article discusses how and why this proliferation has occurred, focusing on the Cambodian context. Highlighting that models are developed comparatively, with reference to one another, the analysis shows how they have generated a partially connected ecology of comparisons. As each model makes its own image of the Mekong, the ecology as a whole creates a kaleidoscopic effect. In principle, this ecology is important for that of environmental policy-making. In practice, however, it is tremendously difficult for scientists to bridge the ecologies. Examining two cases of NGO-based modeling aiming to influence policy, the second half of the paper offers a comparative analysis of the challenges modeling knowledge faces in Mekong environmental politics.