Well, my answer will not go far from 'nothing at all', but maybe it'll be of some use.
GtkD is basically a wrapper around GTK+. That means, that there's almost a one-to-one correspondence between its functions. Also, this means that the object models are exactly the same.
E.g. there are such entities like windows, buttons, etc. Those have some relations, like, button can be placed on a window. Conceptually, all of those objects and relations are exactly the same for both GtkD and GTK+ (v3.x).
And for a GUI toolkit this object model is the most important thing to study and understand. Also, it's the biggest thing. So, if you understand this model: what objects there are and how they can be used together - you'll be successful in writing GUI apps using either GTK+ or GtkD.
The syntactical difference you'll need to overcome is pretty trivial compared to this.
I would suggest to use the GTK+ documentation when looking for how-to-do-stuff and use the GtkD documentation to understand how those concepts are implemented in D. E.g. in GTK+ v2.x to add a button on a window you use something like
gtk_container_add (GTK_CONTAINER (window), button);
and in GtkD you can do the same by
window.add(button);
In other situations it can be a bit more complex and not that straightforward at first, but will be more and more easy when you get some experience.