Currently, research interest is growing regarding integrated modeling of travel behavior considering the effects of activity-duration and trip-scheduling decisions. Capturing interdependencies between these decisions is fundamental to estimate traffic flow in urban transportation networks. Given that integrated models capture simultaneity of these decisions, they are expected to achieve better model realism compared to sequential-based models. However, simultaneous consideration of many behavioral decisions introduces some modeling challenges, including high dimensionality, model uncertainty, complex model specification, and computational intractability. Most of the current literature focus on parametric models that impose strong restrictive assumptions by prespecifying the functional form and number of parameters. It is difficult to know a priori the most appropriate function to use to model complex integrated decisions. To address these modeling challenges, a nonparametric Bayesian approach is proposed to estimate an integrated travel demand model of work duration and commute mode. The proposed nonparametric model, a generalized additive Gaussian process model, precludes the need to prespecify a functional form. The additive structure of the model enables computational tractability in high dimensions of covariates. Bayesian estimation was adopted for inference to quantify uncertainty using probability distributions.