Apologies for the delay in getting back to you about this. 
I have been thinking about your idea: the basic difference between functions and your example is that you presented a declaration of a local variable instead of a constructor calling. Ok, it is a special case because the constructor gets invoked when the flow gets into the scope, but I think this is basically a declaration rather than a function calling.
Despite this I would put in a feature request, but I run into problems. Let's see your example with the class declaration:
class CMyClass
{
CMyClass(int index, int subIndex, bool readOnly);
void DoIt();
};
class COtherClass
{
void Member()
{
CMyClass myClass(0, 1, false);
// lots of code here...
myClass.DoIt();
}
};
myClass in the first and in the last line of "Member" function treated as the same way, so you will see the same tooltip and definition field on both lines.
The first question is it logical? I think it is.
My second question is about the content of the tooltip. Tooltips and definition fields are basically the same, but in this case definition field is smarter: I see this:
CMyClass myClass(0, 1, false)
over myClass in the last line of the "Method", which is good, because you see the whole definition with the incoming parameters.
This is the same as when you define a variable, for example
int apple=5;
and later you can see the type and the default value of apple without the need of Alt+G.
I think tooltips shouldn't behave differently than context and definition fields. That may would confuse users and I think they expect for similar functionality.
I hope this makes sense.
