 At the end of 2020, over 82 million people worldwide were forcibly displaced from their homes, nearly half of them children. Many people and families who have forced to flee must often undertake perilous journeys to reach safety and asylum. The devastating effects of displacement on human life and economy and challenges for governments around the world are clear signals that we need a better understanding of migration and migrant journeys. The authors of a new book from Springer's Methodos series hope to provide the tools for that undertaking. In, towards Bayesian model-based demography, agency, complexity and uncertainty in migration studies, researchers across several disciplines offer a much-needed blueprint for modelling modern migration processes. The authors pull from recent advances in demography, computational modelling, statistics, cognitive psychology and computer science to tackle the uncertainty and complexity that shroud international migration. Their approach relies on agent-based modelling, a choice that reflects a focus on the migrant journey, the information migrants gather on their journey, and how they make decisions based on that information. Embedded in the real spaces across which people migrate, this method provides a way of mapping the dynamics of migrations and how they evolve over time. The authors discuss how to source and integrate various types of data into their modelling framework through an iterative process, taking the Syrian refugee crisis as a timely and compelling case study. As the title of their book suggests, the authors use Bayesian techniques to help readers design ways of quantifying uncertainty and complexity and calibrating agent-based migration models. Statistical models, of course, are only as good as the data on which they're based. Sure migration models reflect the reality of migrant journeys, the authors detail examples of data sources that can inform models of asylum migration. They also discuss new types of psychological experiments that could help researchers address current limitations in gathering empirical evidence. To help execute this new vision for migration modelling, the authors describe their formulation of a dedicated programming language that is flexible, efficient and easy to use. Naturally, many facets of migration studies can't be solved by modelling alone. The human mind and spirit can't be reduced to a set of equations, however sophisticated they might be. But by making informed and transparent choices along the way, modelling provides a useful place to start exploring important questions and learn new ones to ask.