The expected utility theory/hypothesis (EUT/ET) by von Neumann and Morgenstern has become the subject of criticism and efforts to improve it already at the moment of presenting its complete axiomatic form (1953). The broadest analysis was carried out by Kahneman and Tversky using the prospect theory (PT). However, it was not carefully prepared. The above two were helped in 1982 by Quiggin with his extension of the EUT in the form of the rank-dependent expected utility (RDEU). This was followed by Kahneman and Tversky in 1992 boasted the second version of the prospect theory, i.e., the cumulative prospect theory (CPT). It became the most competitive proposal to the EUT, although later other researchers added new elements to it. In this way, the subsequent generations of the PT were created. Today there are five in total. It soon became apparent that the PT also did not satisfactorily explain various cases of decision-making under conditions of risk and uncertainty. Upon closer analysis, however, it turned out that the PT was in fact a generalization of the EUT. Both theories are tools for modeling insurance decisions, including agriculture. In this context, the main aim of the article is to present the principles and the results of applying them in agricultural insurance. The analysis shows that the pragmatic solution at present is the combined use of both theories, with the EUT usually being the reference point.