Huge resources are currently being deployed to survey our Galaxy with the goal of unravelling its structure and assembly history. On account of our peculiar position in the Galaxy, any survey is subject to strong observational biases. The natural way to overcome these biases is to compare the data to pseudo-observations of model galaxies. The potential of different kinds of Galaxy model for this work will be reviewed and the advantages of models based on a new technology, "torus modelling", will be presented. Some early results obtained from torusmodels will be described, and the likely future direction of the field discussed.