In this paper a propeller model is developed for use in computational fluid dynamics (CFD) simulations of ships in waves such that the computation can be performed without having to discretize the propellers. Previous work has developed a propeller model based on a Gaussian Process Regression formulation that predicts the longitudinal and side forces in terms of the ship instantaneous rigid-body motion. That form of the model is trained on static-drift data, and is prone to inaccuracy during transient maneuvers. Also the previous model is not naturally aware of the wave velocities. In this paper new models are proposed that incorporate local fluid velocity information that are shown to enhance accuracy when tested on unseen transient forced yaw maneuvers.