Model Comparison, DiscreteUniform vs. Bernoulli, ElemwiseCategorical vs. CategoricalGibbsMetropolis

LOO is Leave One Out Cross Validation and WAIC is Watanabe–Akaike Information Criterion, correct?
Then you’re basically saying that model bayesian model comparison via a hierarchical model is “out”? Is this because of its technical difficulties (like my problem) or some fundamental theoretical reason?
Thanks!
Christian