In the aftermath of the rational expectations debate and the onslaught of the New Classical economics, some builders of macroeconometric models have begun to change some of their habits, arguably for the better. In particular, neoclassical discipline is increasingly respected in the formulation of the steady states or balanced growth solutions of the latest versions of several models (e.g., Australia's Murphy Model, and the McKibbin-Sachs Global Model). As well, the behaviour of certain variables (especially exchange rates and investment) increasingly tends to be linked to intertemporal optimization. In this paper we evaluate these innovations and illustrate.the role of each, using recent simulations of the Murphy and McMbbin-Sachs models. We conclude that conditions have never been better for convergence in the two streams of economy-wide modelling.