This paper is an exploration into the analysis of public transport networks using space syntax approaches combined with concepts of sustainable accessibility. Present urban development policy aims to achieve sustainable mobility patterns, shifting mobility to soft transportation modes such as walking and cycling, supported by a more integrated multi?modal public transport system, and understanding how different urban areas can support this objective is an important urban policy and design task. We propose that the description of urban areas in terms of ‘modal environments’ is useful to understand the mobility patterns of their populations. We explore this hypothesis by characterising the neighbourhoods in the Randstad region of the Netherlands in terms of public transport ‘modal environments’ by measuring the centrality of the public transport infrastructure of the city?region, and compare their centrality to the mobility patterns of the population. The multi?modal network model of the Randstad includes the three main public transport networks, namely rail, tram and metro/light rail, and it is a simple topological undirected network where the stations or stops represent nodes in the graph and the mobility infrastructure defines the links. In order to create a multimodal network we introduce a fourth set of links representing modal interfaces that make the connection between the nodes of the different public transport modes. We then measure the degree, closeness and betweenness centrality of this network and correlate the results with the data of the 2008 Mobility Survey of the Netherlands, aggregated at neighbourhood level. Although the centrality of the multi?modal network does not fully explain the patterns of public transport use in the different neighbourhoods, it seems to offer a possibility of characterising those neighbourhoods in terms of their public transport use potential, i.e. to what extent they can be public transport ‘modal environments’. The realisation of this potential is dependent on other factors, and further work is required to understand what makes a walking and cycling or a car environment in terms of multi?modal mobility networks. Furthermore, it was found that topological centrality measures show a hierarchy in the multi?modal network infrastructure, and that one should use different methods to analyse the different modes because they play different roles. Finally, it was found that, despite these differences, it is important to consider the network as an integrated multi?layered system. How these layers are modelled and articulated should be subject of further research.