Marine ecosystem models often contain modules for two phytoplankton compartments (flagellates and diatoms) and two zooplankton groups (micro- and mesozooplankton). The models rarely include fish, not even as an agent in zooplankton mortality, which is often formulated as a constant rate. This mortality rate is treated as a free parameter, which can be used to tune or stabilize the model. There are major gaps in our knowledge and modelling capabilities of interactions at the higher trophic levels for example with regards to movement of fish at different scales, prey selection, and zooplankton responses to predators. Here, we argue that there are good reasons for making zooplankton mortality dependent on some key environmental variables known to affect the interaction strength between zooplankton and fish. In addition, since fish are highly mobile organisms, often moving in large groups, there is a need to better understand and model their horizontal migration and to include this in ecosystem models. We present basic models for light-dependent encounters between fish and their zooplankton prey and illustrate how predator–prey interactions can be modelled for herring–Calanus and cod–capelin interactions using individual-based models with super-individuals. In the latter two cases individual displacement is determined by movement behaviour and ocean circulation, and growth and mortality become emergent properties resulting from local encounters between predators and prey. Similarly movement behaviours emerge from simple adaptive rules or more complex models where behavioural strategies are evolved using a genetic algorithm. Such models are versatile and we argue that emergent mortality and growth rates resulting from adaptive behaviours and key environmental forcing are essential for realistic representation of fish–zooplankton interactions.