Good CFA indices but poor SEM indices

Hi all,

I was wondering if anyone could help me make sense of my results from Structural Equation Modelling for my Masters thesis. I've first conducted a CFA and obtained good fit indices, but SEM results have shown poor fit (χ2 (574) = 837.095, p <.001, GFI below .9 and RMSEA was .062. However despite the model being a poor fit, there are some significant relationships between some of my predictors. Can anybody help me understand these results?