This paper presents an analysis of the aftermaths of two firework-blasts from a policy change perspective. The causes of both disasters were completely identical. Both disasters were extensively investigated and findings disseminated. After a 1991 explosion hardly any change in policy occurred while in comparison the 2000 explosion caused gigantic changes, external safety as policy issue developed and became top priority very fast. In this perspective our cases can be considered extreme cases. The analysis is structured by applying six policy change models of which some have already ‘earned’ their footprints in policy-science. The models are the theory of the policy generations (De Vries, 1999), the network- instrument model (Bressers and O’Toole 1999), the stream-model of the policy process (Kingdon 1984, 1995), the advocacy coalition framework (Sabatier and Jenkins-Smith 1999), the punctuated-equilibrium model (Baumgartner and Jones 1993) and the elements of governance model (Bressers and Kuks 2001). The core change mechanisms typical for each of the six models are explicated followed by an assessment whether these can successfully explain stability or change with regard to our cases. The analysis is closed by reflecting on the consequences of the analysis for modelling policy change. Especially the relation between policy sub-unit configurations and core change mechanisms is elaborated.