In practice, urban transformation is often carried out at the outskirts or in between existing areas of a city. In that context, important questions are: How can one imagine a transition between different urban structures? What if, additionally, we aim for including not only the development from innermost to outermost but also vice versa – letting nature into the city rather than the other way around? What are the implications of interventions on the surroundings? Addressing these questions requires a modelling approach that links local morphological compatibility between urban structures with macroscale socio-economic feedback driving development, vacancy and ecological change. To that end, we combine complementary modelling components. A cellular-automaton (CA) grid provides the spatial representation and hosts rule-based land-use transitions over time. Within this structure, an adapted Wave Function Collapse (WFC) operator infers context-sensitive land-use changes from learned adjacency relations, ensuring morphological compatibility of infill and transformation. A system-dynamics (SD) layer generates macro-scale pressures in households, jobs and capital that drive development, vacancy and decline across the grid. Evolution over time emerges from the CA rule system and from vacancy-driven transformations derived from previous work on urban cell grammars, extending WFC from a static pattern generator to a dynamic operator of urban change. Our framework is intended for use in planning, but also in education and policy making (“What-If”-Scenarios). The system supports both conventional development trajectories and their inversion: not only densification and infill, but also decline, abandonment, and ecological succession where nature re-enters under conditions of sustained vacancy.