Based upon the contributions of the 8th ISEM conference on the state-of-the-art in ecological modelling in Kiel (Germany) recent general trends are discussed. The tendencies mentioned refer to theoretical and technical research as well as to environmental applications. Present changes in modelling philosophy have led to an estrangement from the mechanistic paradigm towards a holistic and dynamic approach combining different levels of integrations stressing the self-organization capacity of ecosystems. Due to these properties we have to accept an irreducible extent of uncertainties which characterize ecological developments. This limits model prognoses and requires particular management techniques to deal with unexpected situations. The potential of new modelling techniques to cope with these limitations are discussed. An intensified cooperation is demanded especially in the context of environmental application of models, which should be enforced by initiatives of scientists adressing decision makers as well as societies and human population.