Research has shown that prevalence of childhood stunting in Namibia is currently about 24% [6]. However, there has not been in-depth statistical modelling of childhood stunting done in Namibia. The main objective of this study was to fit a Bayesian additive quantile regression model with structured spatial effects for severe childhood stunting in Namibia. The 2013 Namibia Demographic and Health Survey (DHS) data was used in this study. Statistical inference used in this study was fully Bayesian using R-INLA package. Significant determinants of severe childhood stunting ranged from socio-demographic factors to child and maternal factors. In particular, we found that severely stunted children were those belonging to male headed households, dwelling in rural residences, whose mothers had low education, with frequent exposure to diarrhoea, with HIV+ status, and belonging to poor households, Furthermore, child age and duration of breastfeeding had significant nonlinear effects on severe childhood stunting. We also observed significant positive structured spatial effects on severe childhood stunting only in Ohangwena, Kavango, Hardap, and the Karas regions. We recommend that childhood malnutrition policy makers should consider timely interventions based on risk factors as identified in this paper including spatial targets of interventions. We further recommend that maternity leave be extended to six months to allow optimal breastfeeding especially to mothers with busy work schedule.