This paper takes into consideration the severe bottlenecks that have actually bedeviled econometric analysis and documentation of food security since time immemorial. It aims at modeling food security estimation using fuzzy logics. The paper shows econometrically how food security measurement drawbacks are overcome using residual diagnostic analysis by the effects of fuzzy logics on the leverage points of food security predictors. Further, the results indicate that the preliminary econometrics tests on the residual diagnostic analysis on the error variance, co linearity, multicollinearity and mahalanobis distances improved the estimation of food intake (the predicted criterion) because its predictors are stabilized upon data conversion into fuzzy membership functions. To a certain reasonable extent, it may be very safe to conclude that there is something quite positive in econometric research when fuzzy logics are applied in estimating food security, poverty among other similar subjective or qualitative variables.