The lasting theoretical significance of the South East Asian financial crisis of 1997/98 is that it poignantly offers a critical empirical assessment of equilibrium-centered (first generation) models of international financial crisis. Situating discussion initially in a vignette and commentary on recent events in Argentina, and while reflecting upon the S.E. Asian crisis, we comment upon policy and protocol shifts evident in the International Monetary Fund. These are indicative of a demand for greater rigor in theorizing and modeling of international financial crises. From problematizing first- and second-generation models the paper turns to explore the conceptual basis for, and possible meanings of a third generation of models grounded upon Chaos Theory. At this juncture the domains of prediction and financial crisis instability are explored in formal terms. In this connection, the Lyapunov exponent is presented as an indicator of chaos or crisis instability that may be used to address such issues as the relative (in)stability of the international financial system. This proposed third generation of models could hold the middle ground between the often-formalized first generation of conventional neoclassical economic analysis and generally under-formalized Post-Keynesian second generation.