First don't dismiss SPSS so quickly yet. I'd suggest spending some times with the graph editor (by double clicking on the resultant chart) and mess around with what you can customize. With some proper alignment, color choice, and rescaling the ordinary SPSS graph can look very nice.
For graphing, I will recommend R (http://www.r-project.org
). You can import SPSS data into R with the "foreign" package (R calls all the add-ons "packages," which are freely available on the Internet) and carry out your graph making. Two major graph packages for R you should definitely check are:
Lattice (lmdvr.r-forge.r-project.org/figures/figures.html) and
The links are their official gallery. And they both have an affiliated book published by Springer walking you through the functions.
Even without the packages, R itself has a very versatile graphing mechanism (addictedtor.free.fr/graphiques). Users can basically draw things down to the level of pixel. The drawback is that it's script-based, which mean it's another language to learn, but once you have designed a graph, you can save the syntax for future use. Users with programming background will also find this easy to pick up as well. R also has one of the most active online communities which can definitely help you with your question.
And do you know the best part? It's actually a statistical software! You may even analyze your data with it!
Note that all these software packages (R, SAS, SPSS, Stata) do not really produce the typical "info graphics," which I loosely refer to as a collection of textual and visual displays, often visually pleasing, to organize and show a coherent cluster of information. If you're thinking about something reasonably simple, then exporting the nice images and strategically lay them out on a big poster/presentation slide can usually do the trick. However, if you're thinking about something complicated as this (timeplots.com/potus/?a_aid=coolinfographics), then the best bet is a combination of a stat software and a graphic software like Adobe Illustrator.