Should I use access as a front-end or should I develop a custom front end using vb?
You probably already know the pros and cons of this, but I'll point out the glaringly obvious. Also, I'm assuming you're talking about VB.net here.
Access pros: it's already there and it already works.
cons: Access UIs generally suck. You're constantly working around Access's annoyances. Access uses some dated interfaces. You have to ship the Access runtime.
VB pros: It's more flexible, modern and generally More Gooder. VB has lots of annoyances to work around, but not nearly as many as Access. VB UIs are a little bit better, assuming you're good at writing UI code.
cons: You'd have to redo it from scratch. You have to reimplement a ton of stuff you get for free from Access. You have to ship the .NET runtime.
So... however it stacks up for you. That whole "not redoing everything from scratch" part is a biggie, but I definitely sympathize with you if you're thinking about how irritating Access can be.
I would also consider the scale of it. Not knowing the number of forms but only the queries and tables I would think in terms of ROI. Sc008y is right - you'll have to put in more hours in .NET. Does it justify the time investment? You might also want to research how easily you might replicate the Access reports also - VS.NET does not, as standard, have anything that matches up to Access reports without either more $ investment or more technical tweaking (e.g. IIS & Reporting Services).
Another point - a .NET deployment looks much, much better on the CV\ Resume.