I happen to be Uruguayan and currently live in Uruguay... and while I endorse the OLPC project in my country (Plan Ceibal), I'd say you're grossly exaggerating its results - we already had a very good literacy prior to it (as in, better than the US), and the XO itself might not even have been the cheapest option.
The assessment from the study cited in the article I linked to might exaggerate the results, but note that the reference was to computer literacy, not literacy -- there's a pretty big difference between the two. (And, incidentally, the US has pretty abysmal levels of functional literacy, so if you are using any meaningful measure of literacy to compare with, I wouldn't classify "better than the US" as necessarily "very good" when it comes to literacy.)
and the XO itself might not even have been the cheapest option.
The XO may not be the cheapest (or even, what is more important, cost effective) option, perhaps -- though "might not be" isn't the same as a credible basis for believing that its not -- but certainly, the existence of the XO has spurred lots of efforts by big players in the hardware and software industry whose place would be threatened by widespread adoption of another platform to compete directly with it, and those other companies are defending long-term profit interests that go well beyond the immediate market for the XO and similar devices. But those other choices wouldn't exist if it wasn't for the OLPC project.
And, whether its necessarily the best product or the best solution, its hard to call the XO's that have been delivered, and what they have enabled, not a real product or a real solution.
Though if the OLPC project had not existed, I doubt such a far-reaching and ambitious plan would have been implemented, so even if it was more PR than anything, it WAS important, in making the politician's minds open to the possibility (and it was a HUGE selling point for politicians of the current party in power at the recent elections which they won).