In this paper, a structural model of the demand for multi-destination non-work travel is developed, and an empirical illustration of the structural model is presented. The structural model is based on the principle of utility maximization and differs from previous utility maximizing models of multi-destination non-work travel in several important ways. In contrast to most previous models, the model presented here incorporates travel frequency, destination choice and mode choice for both single and multi-destination travel into a unified utility-maximizing framework. The model includes a representation of the demand for travel between individual origin-destination pairs but avoids the need for enumerating complete travel patterns. Finally, the model incorporates the concept that current travel decisions depend on past travel decisions and future travel plans, as well as on current conditions. Empirical tests of the model have produced encouraging results concerning the model's structural validity. The empirical tests also have indicated that there is a need to develop improved sets of explanatory variables for non-work travel.