This review assesses gaps in water quality modeling, emphasizing opportunities to improve next-generation models that are essential for managing water quality and are integral to meeting goals of scientific and management agencies. In particular, this paper identifies gaps in water quality modeling capabilities that, if addressed, could support assessments, projections, and evaluations of management alternatives to support ecosystem health and human beneficial use of water resources. It covers surface water and groundwater quality modeling, dealing with a broad suite of physical, biogeochemical, and anthropogenic drivers. Modeling capabilities for six constituents (or constituent categories) are explored: water temperature, salinity, nutrients, sediment, geogenic constituents, and contaminants of emerging concern. Each constituent was followed through the coupled atmospheric-hydrologic-human system, with prominent modeling gaps described for a diverse array of relevant inputs, processes, and human activities. Commonly identified modeling gaps primarily fall under three types: (1) model gaps, (2) data gaps, and (3) process understanding gaps. In addition to potential solutions for addressing specific individual modeling limitations, some broad approaches (e.g., enhanced data collection and compilation, machine learning, reduced-complexity modeling) are discussed as ways forward for tackling multiple gaps. This gap analysis establishes a framework of diverse approaches that may support improved process representation, scale, and accuracy of models for a wide range of water quality issues.