If your DP has totally normal sperm then the odds are 50/50 for each one. Assuming that, you already have a 25% chance of having 2 girls anyway.
I'll try to do a diagram (formatting goes funny usually)
==Baby 1 ========== Baby 2
Boy (50%) --Boy (50%)
................\__Girl (50%)
Girl (50%) --Boy (50%)
...............\_Girl (50%)
You can see there are four outcomes, each equally weighted. A boy then a boy, a boy then a girl, a girl then a boy, a girl then a girl. Each outcome has a 25% chance of happening, which is quite high. And that's with totally normal sperm. Of course if your DP happens to have some kind of genetic issue with his Y chromosome, then it will be hard for him to conceive a boy.
Like when you flip a coin, the sex of previous babies doesn't affect the subsequent babies' sex. It's not like you're drawing things from a limited supply - for example if you have five red counters and five blue counters in a bag, and you draw out two red counters in a row. The third counter is more likely to be blue than red because there are still five blue counters in the bag but only three red ones.
I think the 53% status is more accurate. Your base chance is 50/50 but there is a small possibility that one of you has an issue which means that a boy foetus wouldn't be viable. That would probably be less than 3%, but because you've had only girls so far, there's a slightly higher chance you're one of those couples, hence the 3% extra bias. (actually there are slightly more boy babies born than girls anyway, 51/49 so it's a 2% bias)