Probably the most important single improvement you could make is to get a better copy of the original to work on. By this I mean:
(a) Vastly higher pixel dimensions (the one you posted is only 725 px wide);
(b) Much higher spatial resolution (ie, pixels per linear inch of the original);
(c) 16 bit TIF instead of an 8 bpc JPG;
(d) Much brighter and with the histogram spread out much more.
If you scanned or photographed the original, this shouldn't be a problem. However, if you do not have access to the original, then we've got to deal with the image that you posted.
My approach was to take the image into ACR, and adjust it to maximize the brightness of the red areas while keeping the black characters as dark as possible.
As you can see, this has the effect of brightening and spreading out the histogram, as suggested above, but doing so after the fact is no where near as good as doing it either in the scanner or camera.
I then took the image into PS, and made a copy of the red channel. I then did a "Levels" adjustment on it and obtained the following.
The result is a bit different than your attempt -- maybe better in some ways, not in others. This is a classic example of the old saying, "Garbage In, Garbage Out", ie, you need to be working with a better starting image.
HTH,
Tom M