Regardless of the steady increase in the economic growth of South Africa, poverty levels and food insecurity have not declined as one would have anticipated. Urban and peri-urban agriculture (UPA) presents an opportunity as a livelihood strategy to alleviate poverty and ensure household food security within the urban and peri-urban spheres. However, less research has been done in this area to discover the contribution of UPA on households’ food security. This paper assesses the food security status of households that practised any form of UPA activities (later on referred to the rest of this paper as UPA practising households) and those that did not (later on referred to the rest of this paper as non-UPA practising households) within the Tongaat peri-urban area of eThekwini Municipality. The specific objectives of the paper are to estimate the household food security status of the UPA practising households vis-à-vis the non-UPA practising households and to elicit the reasons associated with the practice of UPA activities. Two hundred and eight (208) households (that is 109 and 99 UPA and non-UPA practising households respectively) were selected using a stratified random sampling procedure. The Household Dietary Diversity Score (HDDS) and Household Food Insecurity Access Score (HFIAS) measures were employed to estimate a household’s food security status. A probit regression analysis was used to elicit the reasons associated with the practice of UPA activities by households. A non-parametric independent samples test (Mann-Whitney U) was used to compare whether there were significant differences between the two groups. A Pearson Chi-Square test reveals that the employment status, access to arable land, land tenure (ownership rights or arable land) and household monthly income variables were statistically significantly associated with the food security status (in terms of HDDS) of households. The results from the HDDS tool, show that a greater proportion (54%) of the UPA practising households consumed >6 food groups (deemed to be food secure in terms of dietary access) as compared to their counterparts, the non-UPA practising households (40%) in the same food group. However, the Mann-Whitney U test (U = 5292, p = 0.808) show that there was no significant difference from this data in terms of the dietary diversity (HDDS) of the two groups. The HFIAS measure reveal that a greater proportion (about 72%) of the UPA practising households indicated that they never or rarely worried about food shortages (deemed to be food secure in terms of food access) as compared to their counterparts—the non-UPA practising households (about 61%) that never or rarely worried about food shortages. The Mann-Whitney U test (U = 4118.5, p = 0.001) show that there was a significant difference in terms of food access (HFIAS) of the two groups. Overall, the results show that although UPA practising households seemed to be better off in terms of food access as compared to the non-UPA households, the results are inconclusive or show no evidence that a significant difference existed concerning dietary diversity of the two groups. The probit regression analysis shows that the variable arable land size (p = 0.000) has a positive correlation with the practice of any UPA activity. In order for UPA to enhance the household food security status (particularly in terms of dietary diversity) within the peri-urban spheres, an integrated approach (with agricultural land support from government and city planners) together with the diversification of high-value UPA activities by households is paramount.