I took my own crack at this and you can see my results at:
http://www.glennchan.info/Proofs/dvinfo/jxc/jxc_corrected.jpg

The approach I took:
The first thing I tried was using the color curves filter to take a look at highlight detail. If you make the curve concave up, then that'll "expand" out the highlight detail. The highlights look kind of garish and I really don't know of anything to fix that, unless you want to add diffusion effects or whatever and make it look "artistic".

Anyways...
Sometimes you can rescue additional detail out of the highlights by 'mapping' color information to luminance information. i.e. if you want to rescue detail out of a blue sky, you can map information from the red (and green) color channels to luminance information to give you more detail in there. Ok that's a bad example because you're not going to get much details in skies (except for clouds maybe, but they're white).
That's in theory anyways.
This is best done in Photoshop, which has the luminosity composite mode. There are things in Vegas you can do to approximate this.

In this case, if you take a look at the green channel only, you'll notice it looks the least garish. You can do this by making a bunch of presets in the channel blend filter and cycling through them. Or you can go to the video preview window, the overlays drop down button (looks like # ), and select green as grayscale.

To map the green channel to luminance:
Superimpose the video onto itself. Crtl-drag the clip onto a higher track.
Change the composite mode on the upper track to multiply (the green X and the filmstrip icon).
In the lower track, click the "L" icon (with the up arrow) to establish a parent/child relationship so the multiply mode doesn't affect every other track underneath.
On the top clip, apply the channel blend filter. My settings:
Red = 0.3*R + 0.6*G + 0.1*B (make sure these numbers add up to 1)
same for green and blue
If you undo the parent/child relationship and solo each track, you'll see what's happening.
I chose those particular numbers because if I used only the green channel, things would start looking unnatural.
The multiply mode is nice here because it maps the color information to luminance information, and it darkens the entire shot while pulling out highlight detail (making it more visible).

Using the channel blend filter in this fashion makes only small differences. The dress looks slightly better, and the wood in the chairs looks slightly better. Flesh tones on the guy in the foreground look a little worse, and the cross looks purple-ish.
The colors may look slightly unnatural because this method produces color shifts in terms of hue (look at the vectorscope, the colors move around there). If Vegas had a luminosity composite mode, I don't think this would happen as much??
It may be that the image looks better without the channel blending stuff.

You can download my .veg at
http://www.glennchan.info/Proofs/dvinfo/jxc/
It shows four different ways of approaching this.

2- White balance first. I ultimately do this by eye. Drag the highlight color wheel around, holding crtl to get finer increments. Double check correct WB by cranking saturation.
I didn't balance the midtones because I was being lazy.

3- You will need to add a color corrector filter after everything to convert from 0-255 (computer RGB) color space to 16-235 (studio RGB). DVD players (some, but not all) will clip all values outside the 16-235 range. Use the color corrector filter, and the computer RGB to studio RGB preset. It'll reduce saturation too, which may or may not be what you want.

Using the composite modes means you should use nesting, or render out all the original footage with the corrections applied.

4- You can also add the unsharp mask filter, using the "light" preset. This will kind of make things look sharper.

*Digression: Please make a feature request for a luminosity composite mode. :)