Is it OK to lag first differenced variables?

#1
In a time series regression, where all variables are in first differences, to address the sationarity issue. Would that be OK to lag idependent variables, so that endogeneity problem is addressed or would too much information be lost?

Thanks.