In your both models, mediating effect of SWB is insignificant as the indirect effect is measured as follows; effect size of path a(X>M)*effect size of path b (M>Y). Where as, total effect refers to direct effect+indirect effect. As far as, significance level is total effect is concerned, it will always be significant if the direct effect between X and Y path is significant.

You need to assess the significance of the path a*b where the SE can be calculated manually (eg one easy one is the SE in the denominator of Z in Sobel test). Essentially this is the indirect effect which if signif you have mediation - total if direct is insignif and partial if it remains significant).

In this case a times a small number will produce a small number...but that needs to be ratioed against the error in those two effects a and b.

Hope that helps