Yes, they are mostly getting increasingly complex, but you will still find some pretty useful ones among new data structures developed recently, and also new ways of doing the old stuff in a faster / better / simpler way.

Some examples:

• In 2008 Sedgewick published the Left-leaning red–black tree, a RB-tree with a simpler implementation.
• Counting Bloom filters introduced by Fan et al. in 1998, improved versions are still being developed.
  
• Cuckoo hashing by Pagh and Rodler in 2001.
• A rather simple (and fast in practice) O(n) construction of the Suffix array and LCP array by Nong, Zhang, and Chan in 2009.
• A lot of work has been done in the recent years on Persistent data structures (i.e., ones that can be queried about their contents in the past) and on Succinct data structures (i.e., data structures such that the total space used by the data structure is close to the information-theoretic lower bound)