The policy preference function (PPF) approach continues to be the subject of considerable interest in agricultural economics. Recent work has added sophistication and strengthened the approach's theoretical underpinnings. In this paper, several implications of this recent work are considered. First, the distinction between the PPF and the surplus transformation curve (STC) is stressed. Estimated PPF weights are derived from what we know about the STC. We actually know very little about the PPF itself. Second, the relationship between the number of interest groups and the number of policy tools in PPF models is discussed. The relationship that is mathematically necessary may not correspond to that which is suggested by our intuition about reality. Hence, it may be that PPF modelling is often inappropriate. Finally, the stochastic nature of PPF estimates is discussed, and a simple method for quantifying the variability of these estimates is illustrated.